ORACLE/RAC2008. 12. 12. 21:42
반응형
 
★ Check the Hardware Requirements
 /usr/sbin/dmesg | grep "Physical:"
 /usr/sbin/swapinfo -a
 bdf /tmp   
 bdf
 /bin/getconf KERNEL_BITS
 uname -a

★Async
To verify Async I/O is running, do the following procedure with gpm or glance.
     1. Select the dbwr process from the process menu.
     2. After selecting the process, select the "open files" option
 from the menu.
     3. Look for the /dev/async file.
Glance reports which files a given process opens. When async I/O is used,
it opens /dev/async as a file. Rememeber you also must set the
 disk_async_io = TRUE parameter in init.ora parameter file to use async I/O.
 
★ IP Address Requirements
 -------------------------------------------------------------------
 Host Name               Type       IP Address       Registered In 
 -------------------------------------------------------------------
 rac1.mydomain.com       Public      143.47.43.100    DNS 
 rac1-2.mydomain.com     Public      143.46.51.101    DNS 
 rac1-vip.mydomain.com   Virtual     143.46.43.104    DNS 
 rac1-priv1              Private     10.0.0.1         /etc/hosts 
 rac1-priv2              Private     10.0.0.2         /etc/hosts 

★ Checking the Network Requirements
 /usr/bin/netstat -in
 
★ Check for Required Software
/usr/sbin/swlist -l level | more
  *  -l level         levels of objects to view, one of: "depot",
                      "bundle", "product", "subproduct", "fileset",
                      "file", "root", "shroot", "prroot",
                      "category", "patch", "control_file"
 
★Check for Required Patches
 /usr/sbin/swlist -l bundle | grep GOLD
 /usr/sbin/swlist -l patch | grep PHSS_28880
 /usr/sbin/swlist -l patch | more
 
★ Creating the OSDBA Group
 /usr/sbin/groupadd dba
 
★ Create a New Oracle Software Owner User
 [node1]
   /usr/sbin/useradd -g dba oracle
   passwd oracle
   id oracle
     -> uid=440(oracle) gid=200(dba)
 [node2]
   /usr/sbin/groupadd -g 200 dba
   /usr/sbin/useradd -u 200 -g dba oracle
   passwd oracle

★ Grant Privileges to the OSDBA Group
  /etc/privgroup (<- dba MLOCK RTPRIO RTSCHED 추가)
  /usr/sbin/setprivgrp -f /etc/privgroup (Enter the following command to grant the privileges to the OSDBA group: )
  /usr/sbin/getprivgrp dba
 
★ Configure Kernel Parameters 
  - Bourne, Bash, or Korn shell: 
    $ DISPLAY=local_host:0.0 ; export DISPLAY
  - C shell:
    $ setenv DISPLAY local_host:0.0
   
 /usr/sbin/sam

★ Identifying an Existing Oracle Base Directory
  - Identify an existing Oracle Inventory directory
    more /var/opt/oracle/oraInst.loc
   
  - Identify existing Oracle home directories
     more /etc/oratab
     
    
★ Creating a New Oracle Base Directory    
  mkdir -p /mount_point/app/oracle_sw_owner
  chown -R oracle:oinstall /mount_point/app
  chmod -R 775 /mount_point/app
  
 
★ Create the DBCA Raw Device Mapping File
  find /dev/vg_name -user oracle -name 'r*' -print  dbname_raw.conf
  Edit the dbname_raw.conf file in any text editor to create a file similar to the following:
  - Note:
     The following example shows a sample mapping file for a two-instance RAC cluster. 
 system=/dev/vg_name/rdbname_system_raw_500m
 sysaux=/dev/vg_name/rdbname_sysaux_raw_800m
 example=/dev/vg_name/rdbname_example_raw_160m
 users=/dev/vg_name/rdbname_users_raw_120m
 temp=/dev/vg_name/rdbname_temp_raw_250m
 undotbs1=/dev/vg_name/rdbname_undotbs1_raw_500m
 undotbs2=/dev/vg_name/rdbname_undotbs2_raw_500m
 redo1_1=/dev/vg_name/rdbname_redo1_1_raw_120m
 redo1_2=/dev/vg_name/rdbname_redo1_2_raw_120m
 redo2_1=/dev/vg_name/rdbname_redo2_1_raw_120m
 redo2_2=/dev/vg_name/rdbname_redo2_2_raw_120m
 control1=/dev/vg_name/rdbname_control1_raw_110m
 control2=/dev/vg_name/rdbname_control2_raw_110m
 spfile=/dev/vg_name/rdbname_spfile_raw_5m
 pwdfile=/dev/vg_name/rdbname_pwdfile_raw_5m
 
 
★ Verify the Cluster Software Configuration
 To verify that the HP Serviceguard software is running:
 
 If necessary, configure (or reconfigure) the cluster.
 
 Verify that HP Serviceguard is configured to start automatically at runlevel 3.
 
 To verify that HP Serviceguard is running correctly, enter the following command: 
 
 # /usr/sbin/cmviewcl -v
 
 Verify that the cluster status is up, and that all nodes are up and running.
 
 If the cluster is not running, enter the following command to start it:
 
 # /usr/sbin/cmruncl
 
 If any nodes are not running, enter a command similar to the following to start the cluster daemon on those nodes:
 
 # /usr/sbin/cmrunnode nodename1 nodename2...
 
 If you intend to use Hyper Messaging protocol as the cluster interconnect,
 enter the following command to verify that the HyperFabric software is configured properly:
 
 # /opt/clic/bin/clic_stat -d NET
 

★ Configure the oracle User's Environment
 $ xhost +  
 $ su - oracle  
 $ echo $SHELL 
   - Bourne shell (sh), Bash shell (bash), or Korn shell (ksh):  
     $ vi .profile 
     $ DISPLAY=local_host:0.0 ; export DISPLAY
    
   - C shell (csh or tcsh):  
     % vi .login
        $ setenv DISPLAY local_host:0.0

 su - root
 # mkdir /mount_point/tmp
 # chmod a+wr /mount_point/tmp
 # exit 
 Enter commands similar to the following to set the TEMP and TMPDIR environment variables:  
 -Bourne, Bash, or Korn shell:  
   $ TEMP=/mount_point/tmp
   $ TMPDIR=/mount_point/tmp
   $ export TEMP TMPDIR 
 - C shell:  
   $ setenv TEMP /mount_point/tmp
   $ setenv TMPDIR /mount_point/tmp
  
    $ unset ORACLE_HOME
    $ unset TNS_ADMIN
    $ umask
    $ env | more

반응형
Posted by [PineTree]